unofficial mirror of libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Wilco Dijkstra via Libc-alpha <libc-alpha@sourceware.org>
To: "naohirot@fujitsu.com" <naohirot@fujitsu.com>
Cc: 'GNU C Library' <libc-alpha@sourceware.org>
Subject: Re: [PATCH v3 5/5] AArch64: Improve A64FX memset
Date: Thu, 19 Aug 2021 13:06:30 +0000	[thread overview]
Message-ID: <VE1PR08MB55991F1636D9A6F5F59E172883C09@VE1PR08MB5599.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <TYAPR01MB6025D4C1DAF98842629CF281DFFE9@TYAPR01MB6025.jpnprd01.prod.outlook.com>

Hi Naohiro,

> In my test environment, Patch V4 has a little bit lager 16KB dip than
> Patch V3 [1].

That's odd since nothing significant changed in the loop. The performance 
variations don't make much sense to me, it seems A64FX is very sensitive
to something, but it's not clear what exactly it is.

Note that I didn't see anything similar for 16KB in bench-memset-large or
bench-memset-walk. However if you'd like to get to the bottom of this, one
way would be to write a small benchmark that reproduces the dip (which is
likely non-trivial) and use perf stat to get performance counter results for the
different loops.

> So let me ask you the same question I asked in the mail [2] before.
> 
> How did you measure the performance? Did you use some kind of simulator?
> 
> Or do you have real A64FX environment?

Yes, I just ran it on an A64FX machine.

Cheers,
Wilco

  reply	other threads:[~2021-08-19 13:07 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-22 16:04 [PATCH v3 5/5] AArch64: Improve A64FX memset Wilco Dijkstra via Libc-alpha
2021-08-03 11:22 ` naohirot--- via Libc-alpha
2021-08-09 14:52   ` Wilco Dijkstra via Libc-alpha
2021-08-17  6:40     ` naohirot--- via Libc-alpha
2021-08-19 13:06       ` Wilco Dijkstra via Libc-alpha [this message]
2021-08-20  5:41         ` naohirot--- via Libc-alpha
2021-08-23 16:50           ` Wilco Dijkstra via Libc-alpha
2021-08-24  7:56             ` naohirot--- via Libc-alpha
2021-08-24  8:07               ` naohirot--- via Libc-alpha
2021-08-24 15:46               ` Wilco Dijkstra via Libc-alpha
2021-08-26  1:44                 ` naohirot--- via Libc-alpha
2021-08-26 14:13                   ` Wilco Dijkstra via Libc-alpha
2021-08-27  5:05                     ` naohirot--- via Libc-alpha

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/libc/involved.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=VE1PR08MB55991F1636D9A6F5F59E172883C09@VE1PR08MB5599.eurprd08.prod.outlook.com \
    --to=libc-alpha@sourceware.org \
    --cc=Wilco.Dijkstra@arm.com \
    --cc=naohirot@fujitsu.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).